Image Segmentation with Deep ConvNets (PyTorch)

Author: Caihao Cui

Lastest Update Time: 28-Mar-2021

Created Time: Sat 27-Mar-2021

This repo is on solving image segmentation task by building a convolutional neural network (UNet) to output a pixel-wise mask of the input image.

The basic function will be like: $ Y_{H, W} = Model(X_{3, H, W})$

where $X$ is a RGB image, $Y$ is a mask image with interger represent one type of objects.

Find more background on the README.md

TOC:

  1. Data Review
  2. Sample Preparation
  3. Build Model
  4. Manage Model
  5. Prediction
  6. TBC.

Data_Review

Sample Visulization

Pick a few image to see the RGB and MASK

Build Model

Manage Model

Prediction

Use the source code

Use Command Line

Read the ImagePair from the path

Sample

Sample

Next Step